mobile feed weather

return to article: "2 very good reasons.."

Aspects of Applied Biology 77, 2006

International advances in pesticide application 2006 

Considering the use of air induction nozzles to reduce operatorcontamination when using knapsack sprayers


1 Harper Adams University College, Newport, Shropshire TF10 8NB, UK

2 Billericay Farm Services Ltd, Downham, Billericay, Essex CM11 1QU, UK


Knapsack sprayers are used throughout the world for the application of chemicals in both agricultural, horticultural and amenity situations. Operator contamination is a concern when using knapsack sprayers. This paper compares low volume air induction nozzles, Billericay Farm Services Ltd. 01 (Orange) and medium volume 03 (Blue) air induction nozzles with more commonly used nozzles of different types to quantify if there is potential for reducing operator contamination. Replicated trials were carried out using all nozzles at 2.0 and 3.0 bar with both minimal wind and moderate wind (average 2 m s-1). Results show a significant reduction in operator contamination when using air induction

Key words: Knapsack sprayers, operator contamination, air induction nozzles


Operator contamination as a result of spray drift has been a major concern with large boom type sprayers and over the years there have been a number of investigations into the creation and movement of spray drift away from the intended target. The effects and quantity of spray drift from knapsack sprayers have not been looked at in as great detail, however, it does occur and it is a source of operator contamination (Wicke et al., 1999). The application volume during hand held applications will affect the severity of operator contamination. Low volumes have traditionally been associated with smaller droplets containing higher concentrations of pesticide leading to greater dermal exposures. However, Moreira et al. (2000) found that the application of higher volumes created a greater risk of dermal exposure when applied under greenhouse conditions.

A new 01 (orange) air induction nozzle offers the possibility of low volume application rates in the order of 55 L ha-1 at an operating pressure of 2.0 bar and a walking speed of 3.6 km h-1 whilst also reducing operator contamination.

Materials and Methods

The sprayer used was an Azo compressed CO2 knapsack unit. This was alternately fitted with an 01 BubbleJet, 01 flat fan, 01 deflector, 02 hollow cone and 03 BubbleJet. Table 1 shows the nozzles used, their flow rates and band widths as measured. Replicates of each nozzle were completely randomised throughout the experiments to reduce the effects of changing weather conditions.

The trials were carried out on short grass (5-10 cm) to simulate the type of spraying which is carried out within the amenity sector in the U.K., for example on golf courses and sports fields. The trials were carried out over two days, both overcast with the first day having minimal wind speed at a height of 2.0 m above the ground with a temperature of 13.4°C and a relative humidity (r.h.) of 95.8%. The second day had a temperature of 9.8°C, wind speed averaging 2.0 m s-1 and an r.h. of 96%. A track measuring 50 m in length was set up parallel with the wind direction, with wind speed and time taken recorded for each replicate.

Table 1: Nozzle flow rates and band widths as measured

  2.0 bar 2.0 bar 3.0 bar 3.0 bar
Nozzle Flow rate / Band width / Flow rate / Band width
  (L min -1) (cm) (L min -1) (cm)
01 Bubble Jet 0.33 100 0.40 120
01 flat fan 0.33 98 0.41 130
01 deflector 0.34 98 0.42 134
02 hollow cone . 0.36 54 0.57 76
03 Bubble Jet 0.62 110 1.19 127

The sprayer was filled wth a mixture of Tinopal (100 mg L-1) and Fluorescene sodium (100 mg L-1) to form a fluorescent spray trace. Wearing a white “Tyvek” spray suit and holding the nozzle 50 cm above the ground using a length of string as a guide to maintain height, the operator carriedout each replicate walking into the wind at a speed of approximately L ms-1. Once the run was completed the spray suits were photographed in the dark under ultra-violet light to show where the spray had accumulated on the operator. The spray suits were then left to dry in the dark for 1 h and were then stored in black plastic bags to prevent photo-degradation of the ultra-violet dye.This process was repeated for each of the five nozzles, with three replicates made for each. For all five nozzles, on both days, the experiment was carried out at pressures of 2.0 and 3.0 bar to identify and compare if an increase in pressure had any effect on the level of contamination from any of the nozzles.

Suits were then cut up into sections of; head,body front,body back,arms, upper leg (knee-waist) left and right, lower leg (knee and below) left and right as carried out by Merritt (1989). Wash off solution made from water with 2 mL L-1 of 1 molar NaOH and 1 mL L-1 Activator 90 (alkylphenylhydroxypolyoxethylene). Sections were then placed into bags with 200 ml of wash off solution and were left for 1 h being agitated regularly to aid complete extraction of the fluorescene dye from the sections. The resultant solutions from each section were placed into containers ready to be put through a Perkins-Elmer LS30 spectrometer for analysis. A standard concentration of fluorescent dye was made up by placing 100μl of spray tank solution into 200 mL of wash off solution. This was used to calibrate the spectrometer to allow all sample measurements to be given in μl of solution on each section.

Data was analysed by ANOVA using Genstat for Windows, version six.

Results and Discussion

The results for operator contamination are shown in Table 2 as the mean operator contamination in μl.l-1 of spray applied for each nozzle. This allows direct comparison between all five nozzles irrespective of volume applied. Table 2 shows that there are significant differences in total operator contamination between the different nozzles under all conditions. At 2.0 bar pressure under minimal wind conditions the two air induction nozzles gave significantly (P = < 0.001) lower contamination than the other three nozzles, however there was no significant differences between the two air induction nozzles. The deflector nozzle under these conditions significantly (P = < 0.001) gave the highest level of contamination, 3179 μl L-1 more than the 03 air induction. At 3.0 bar under the same wind conditions results were very similar with the two air induction nozzles giving significantly (P < 0.001) less contamination than the other nozzles and the deflector nozzle contaminating the operator significantly (P < 0.001) more than the other nozzles.

Table 2: Total mean operator contamination (μl.l-1 applied) for each nozzle at 2.0 and 3.0 bar in minimal wind and breezy conditions*

Table 2: Total mean operator contamination (μl.l-1 applied) for each nozzle at 2.0 and 3.0 bar in minimal wind and breezy conditions* (12Kb GIF)

*Transformed data in parenthesis (square root).

However there was no significant difference between the 01 air induction and hollow cone nozzles although the contamination from the hollow cone was 566 μl L-1 more. Under recommended wind speed conditions for spraying (approx 2 m s-1) the two air induction nozzles gave significantly (P < 0.001) lower levels of contamination than the flat fan, deflector and hollow cone nozzles which
showed no significant difference between them. The 01 air induction nozzle gave significantly (P < 0.001) higher levels of contamination than the 03 air induction, 2822 μl L-1 more. At 3.0 bar in recommended spraying conditions the air induction nozzles are not significantly different but give significantly (P < 0.001) less contamination than the other nozzles. Under these conditions the flat fan and deflector nozzle showed no significant differences between them, giving significantly (P < 0.001) higher levels of contamination than the other nozzles. As pressure increases it is normally expected that the number of small droplets produced by a nozzle also increase and as contamination in this case is a result of drift from the nozzle it could be expected that the level of contamination would increase. Table 2 shows that under minimal wind conditions this is not the case for all of the nozzles, in fact the 02 hollow cone and 03 air induction nozzles showed a decrease in the amount of operator contamination. This also occurred with the 01 air induction, deflector and hollow cone nozzles when applications were made under recommended wind speed conditions.

Fig. 1 shows where the majority of operator contamination occurred for the five nozzles operated at 2 bar under ideal spray conditions. This data is presented as it represents the most likely pressure and wind conditions experienced by operators when spraying using a lever operated knapsack under UK conditions. The lower legs received the highest amount of contamination, indeed this was the case under all conditions tested.

Fig. 1. μL of dye recovered from the nine spray suit sections for five nozzles operated at 2 bar under ideal spray conditions.

Fig. 1. μL of dye recovered from the nine spray suit sections for five nozzles operated at 2 bar under ideal spray conditions. (14Kb JPEG)

The air induction nozzles were shown to produce the lowest levels of operator contamination compared to a flat fan, deflector and a hollow cone of a similar size. The 03 air induction result was particularly interesting as it applies three times as much liquid per minute as the 01 nozzle yet, it was able to produce the same if not less contamination on the operator as the 01 air induction. This
highlights the significant benefits of using an air induction nozzle in that they are able to produce a larger droplet size, therefore reducing the number of small driftable droplets and yet apply the same amount of liquid as a similar sized flat fan. With a trend towards lower water volumes (Jørgensen & Taylor, 2002) it is useful to know that the 01 air induction is capable of producing similarly low levels of contamination to the 03 air induction. Due to the smaller orifice size and lower flow rates it would normally be expected that more small droplets would be produced resulting in more drift and consequently operator contamination even though it is an air induction nozzle. However previous work by Tuck (2003) has shown that the 01 air induction produces droplets with a relatively large VMD of approximately 360 μ at 2 bar and 310 μ at 3 bar, despite its small orifice size. This is similar to the 03 air induction which produces droplets with a VMD of approximately 370 μ at 2 bar and 290 μ at 3 bar. The fact that the VMD is lower for the 01 air induction at two bar could in part explain why the 01 air induction produced a higher amount of total contamination than the 03 air induction in recommended wind speed conditions. However, it is more likely that some other factor played a part in the results shown under recommended wind speed conditions. The most likely explanation is variability in the movement and speed of the wind during the test Wind speeds were on average higher for the runs carried out with the 01 air induction. With a very low proportion of droplets at or below the 100 μm diameter driftable limit, (Jensen & Arvidsson, 2000) the amount of drift and operator contamination is considerably reduced. Miller et al. (2003) stated that the Billaricay air induction nozzle has been shown to consistently produce a relatively small droplet size distribution, and this helps to explain why the two air induction nozzles tested showed no significant differences between them except at 2 bar and in recommended wind speed conditions as mentioned above.

Some of the nozzles tested, in particular the 01 deflector and hollow cone managed to obtain higher amounts of contamination on the operator’s front and back, although these were not always significantly higher than other nozzles. Contamination of the upper body with the hollow cone in particular may have been caused by the way in which the nozzle works. Due to the swirling motion of the spray pattern as it leaves the nozzle, the spray droplets can be moving in a variety of directions and towards the operator’s feet. They are also likely to stay in the air for longer due to lower droplet velocities despite having a low angle of trajectory. This means that it is possible that a number of the smaller droplets are thrown up into the air or are caught in upward air movements and therefore landing on the upper body. However the deflector nozzle does not produce droplets in this way and so it is unlikely that this is the reason for the contamination. Contamination from leaks or dirty equipment is always a danger and can obscure experimental values as the detection levels are pushed lower. The high values obtained from some of the replicates certainly suggest that this maybe the case. This highlights the major contribution spillages and dirty equipment can make to the levels of operator contamination and goes to show how important it is to maintain spray equipment properly.

Contamination was found to some degree on nearly all of the body front and back sections, arms and most of the suit hoods. This was anticipated for the experiments carried out under windy conditions however not during the tests carried out in minimal wind conditions. Despite both experiments providing upper body contamination the results obtained showed that there was less contamination on the upper body sections from the experiments carried out under recommended wind speed conditions. The reason for this could be air currents created around the operator as they are walking. The aerodynamic properties of a spray operator will cause air to move round and over the body. This could create an area of up draught from the front drawing finer spray particles up and onto the front and hood of the spray operator. Air travelling around the sides of the body may result in an area of low pressure air being created behind the operator. Air would then move in to equalise this pressure difference from above and below, possibly in a swirling motion, and as a result draw smaller spray particles up behind the operator causing contamination. Under windy conditions the contamination may be less because the smaller spray droplets, even though still being drawn upwards, would be blown past the operator at a higher velocity if walking into the wind, giving them less time to settle and contaminate the operator. The significance of deposits on head, abdominal and upper leg sites is that penetration of the chemical would be greater in these regions (Sutherland et al., 1990).


This work identifies an improvement in operator safety when applying pesticides using hand held equipment. Where low water volumes are required, an 01 air induction nozzle may be used in preference to other nozzle types that will give higher levels of operator contamination. This may not be of such great importance in developed countries where there is routine use of PPE to protect
the operator, although any engineering method of reducing contamination is of great benefit. More work is needed to investigate the effect of air turbulence around a knapsack sprayer operator on contamination of the upper body.


The authors wish to thank Billericay Farm Services Ltd, Mark Howard-Smith and Matthew Shaw for their help and support during this work.


Jensen P K, Arvidsson T. 2000. Does droplet size affect airbourne drift and sedimentation drift to the same extent? Aspects of Applied Biology 57, Pesticide application, pp. 91–96.
Jørgensen M K, Taylor B. 2002. Knapsack Sprayers. In Hardi International Application Technology Course 1:56–68.
Merritt C R. 1989. Evaluation of operator exposure and spray displacement with hand operated herbicide application systems. In British Crop Protection Council, Proceedings, Brighton Crop Protection Conference- Weeds – 1989. pp.623–630 Farnham: British Crop Protection Council.
Miller P C H, Powell E S, Orson J H, Kudsk P, Mathiassen S. 2003. Defining the size of target for air induction nozzles. Home Grown Cereals Authority Project Report No. 317. London: HGCA.
Moreira J F, Santos J, Glass C R. 2000. A comparative study of the potential dermal exposure of an operator with two pesticide application techniques in a tomato greenhouse. Aspects of Applied Biology 57, Pesticide application, pp. 399–404.
Sutherland J A, King W J, Dobson H M, Ingram W R, Attique M R, Sanjrani W. 1990. Effect of application volume and method on spray operator contamination by insecticide during cotton spraying. Crop Protection 9:343–350.
Tuck C. 2003. Silsoe Research Institute. (Unpublished).
Wicke H, Bäcker G, Frießleben R. 1999. Comparison of sprayer operator exposure during orchard spraying with hand held equipment fitted with standard and air injector nozzles. Crop Protection,

return to article: "2 very good reasons.."